Sensory cilia: How molecular motors can shape ciliary tip organization and diversity

Cilia are organelles that are widely conserved among eukaryotes. They are defined by their architecture made up of microtubule doublets, which are built on a basal body derived from the centrosome. Cilia perform a variety of functions, and ciliary dysfunction in humans can lead to diseases with a wide range of symptoms.

Since their discovery, cilia have drawn considerable attention due to the remarkable diversity of their architectures and morphologies across and within organisms. This diversity is particularly evident in sensory neurons, where sensory cilia exhibit a wide range of shapes and distinct specialized domains. Such structural variation reflects precise evolutionary adaptations that optimize the detection and transduction of diverse environmental stimuli, including light, odorants, chemical signals, and mechanical forces (1).

These sensory modalities are fundamental for animal behavior and enable appropriate responses to environmental cues. In humans, for instance, photoreceptors and olfactory neurons display highly specialized structural and functional adaptations that support their distinct sensory roles. Elucidating the mechanisms that govern the formation and diversification of these sensory cilia represents a central question in cell biology, one that remains complex and challenging to fully resolve.

Among the animal models available to address this question, two invertebrate models, Caenorhabditis elegans and Drosophila melanogaster, have proven to be particularly effective in providing key insights into the shaping of sensory cilia. In C. elegans, ciliated sensory neurons are arranged in a stereotypical manner throughout the organism, with each neuron harboring specific endings that have been thoroughly characterized. Those in Drosophila have also been extensively described at the cellular level. Although specific transcriptional programs have been identified in both organisms to drive the differentiation of specific ciliary subtypes (2, 3), establishing a list of target terminal differentiation genes was not sufficient to explain how each sensory cilium ending is shaped. Nevertheless, an elegant genetic analysis in C. elegans revealed the specific role of a transmembrane protein (OIG-8) in shaping the terminal endings of a subtype of the worm’s olfactory neurons in a dose-dependent manner, with an increase in the protein enhancing the level of ramification of the ciliated endings (4). This mechanism has not yet been proposed in other organisms.

Among the possible molecular pathways that regulate ciliary diversity, the intraflagellar transport (IFT) is a privileged candidate. This pathway is involved in building almost all cilia. It is responsible for conveying cargoes required to assemble and maintain the cilium (5). This transport process involves kinesin and dynein motors, and disturbances to this process can result in the complete absence of ciliary assembly, or more subtle signaling defects in milder conditions. Given its critical role in cilium assembly, it was anticipated that this transport machinery would also be involved in the formation of cilium-specific features. This hypothesis was strongly supported by initial seminal work in C. elegans, showing that different kinesin motors moving along the cilium shape the ciliary endings of its sensory neurons (6). While heterotrimeric kinesin II drives the anterograde movement of IFT in all organisms studied so far, a second homodimeric kinesin, OSM-3/KIF17, has been identified as essential for building the distal segment, composed of singlet microtubules, in a specific subtype of sensory cilia in the worm. Similarly, in vertebrates, KIF17 has been shown to be necessary for forming the outer segment of photoreceptors (7). Interestingly, it has been shown that tubulin glutamylation regulates these IFT motors in a cell type–specific manner to build cilium-specific shapes (8). In addition, a balance in phosphorylation and dephosphorylation events of OSM-3 regulates its activity while being transported by kinesin II to the distal segment to fulfill its function (9). Taken together, these observations suggest that modulating IFT motors may be sufficient to sustain diverse organization of ciliary endings.

In their study, Wang et al. use Drosophila to understand the mechanisms that not only shape the sensory ciliary endings but also organize the molecular domains within them. In Drosophila, the campaniform sensilla are mechanosensory organs that respond to proprioceptive and exteroceptive stimuli essential for locomotion and flight. The ciliary endings of these neurons exhibit stereotyped shapes, featuring a very short cilium and an outer segment comprising bundles of microtubules (tubular bundle), with a distal part enriched in mechanoreceptors, called NompC. This distal segment is known as the mechanosensory organelle (MO) (see Fig. 1). Strikingly, NompC molecules exhibit inside this segment a bimodal distribution, being enriched in the lateral domain and excluded from the tip of the MO. This raises the question of how such a nanodomain of NompC receptors is built and maintained in the MO. One possible candidate involved in concentrating the NompC mechanoreceptors in MO could be IFT kinesins, as suggested by observations in C. elegans. However, subunits of fly kinesin II were found to be excluded from the MO. To identify other candidates, a transcriptomic analysis was undertaken. To this end, the RNA profile- of the two proximal segments of the haltere (scabellum and pedicellus) enriched in campaniform mechanosensilla was compared with those of two controls lacking these sensilla: a third segment of the haltere (capitellum) and brain. Among the differentially expressed genes, they identified the molecular motor Kif19A. They showed that Kif19A is specifically expressed in sensory neurons in fly and precisely located in the MO of campaniform sensilla. Moreover, deletion of Kif19A impaired flight performance and gustatory behavior, demonstrating that Kif19A is required for proper mechanical and chemical perception. The flight behavioral deficit is associated with a distorted distribution of NompC without perturbation of the overall ciliary structure. Loss of Kif19A also affects the doublecortin domain-containing protein DCX-EMAP, which was shown to stabilize microtubules in the MO (10). This function of Kif19A relies on the integrity of its motor domain. Together, these findings demonstrate that Kif19A is essential for organizing the distal specialized domain of the ciliated sensory endings in the MO.

But what specific properties of Kif19 provide this capacity to organize this subsegment? Using in vitro single-molecule motility assays, the authors characterized the motility of the purified Kif19 protein on microtubules. Surprisingly, isolated Kif19A molecules appeared to be nonprocessive, meaning that the motor did not exhibit directional movement on microtubules. However, when multiple kinesins were immobilized on a glass surface at increasing concentrations and incubated with purified microtubules, the latter started to move in a processive manner as the density of Kif19A increased. This indicates that this motor could shift from a nonprocessive motile force to collectively driven processive movement toward the microtubule plus end. Using these single-molecule microtubule-binding assays, the authors show that Kif19A indeed transports NompC and DCX-EMAP in a processive manner along microtubules.

Is this characteristic sufficient to drive the molecular nanodomain organization of the distal MO? To answer this question, the authors used mathematical modeling and took into account the motility parameters of Kif19A and estimated diffusion rates of soluble and membrane proteins in the MO. This allowed them to simulate the distribution of the NompC mechanoreceptors inside the MO. A key feature of this model is its assumption that NompC diffusion is restricted by signals from neighboring cells. Indeed, previous work has demonstrated that the removal of NompA in the adjacent cell results in a loss of NompC’s polarized distribution (10), thereby supporting this assumption. Using this mathematical modeling, the authors demonstrated in their model that a precise balance between diffusion, Kif19A-mediated directed transport, and local binding that restricts NompC diffusion could result in the bimodal distribution of NompC, as observed in vivo and supported by a Kif19A KO phenotype.

The work of Wang et al. (11) hence demonstrates that local directed transport by molecular motors contributes to the molecular organization of ciliary endings. The proposed model also helps us to understand how the diversity of molecular domain organization in sensory ciliary endings could be achieved by manipulating any one of the three parameters described here: directed transport, free diffusion, and local binding. This work also raises many captivating questions: how is Kif19A enriched at the MO segment and are IFT kinesin II motors excluded from it? Does Kif19A enrichment depend on IFT, and how does Kif19A transport interact with the IFT machinery? When Kif19A is removed, the organization of the molecular nanodomains is altered, but the overall shape of the distal segment is conserved. What are the other elements that contribute to shaping the MO? All of the questions state that further work is required to refine the proposed model.

In conclusion, this research provides critical insights into the mechanisms that govern the molecular regionalization of sensory cilia, a process essential for the behavior of many organisms.